A Look Beyond Basic Production Records


By Dr. Tom Gillespie, DVM,
Rensselaer Swine Services

Records systems have been available for more than twenty-five years, with the first activity starting when PigCHAMP© became available in the mid 80s. Record systems help producers track every possible production parameter in an effort to measure, monitor, and ultimately manage these parameters. Routine review of production records can help identify changes before they become clinically obvious or at least obvious enough to warrant a phone call to the attending veterinarian. In addition, records have been a tremendous help in analyzing and understanding the production challenges that are faced after a clinical disease outbreak.

Production records can also be used in the incorporation of partial budgeting activities, to further understand the cost of disease. This ensures proper use of available tools, such as medications, vaccinations, management strategies, etc., can be utilized economically by veterinarians and producers. Production record accessibility is becoming more streamlined with the accessibility of online access. The capture of information at the farm level can be uploaded and production parameters examined within a short period of time with online access to programs.


Many reports that are available to us provide tremendous amounts of data, but in a random unorganized fashion – this makes analysis more difficult for most of us. Despite software and hardware improvements over the years, productivity monitoring can be a bit of a nightmare when an in-depth analysis is needed. Production managers and owners rely on the numbers to indicate that they are “going in the right direction”. If the numbers change for the better, compared to last month, one usually feels good. If the numbers actually change for the worse compared to last month, especially if the comparison was not expected, then the overall feeling of doom settles in. Owners despair and production and site managers consult with their staff to try to determine the causes of downturn. Veterinarians may be consulted. The main challenge in understanding production records is separation of what has become known as “biological noise” versus a definite trend that requires energy and attention to comprehend.


One of the problems in our current production era, was summarized by Daniel Boorstin when he said: “Information is random and miscellaneous, but knowledge is orderly and cumulative.” In short, raw data has to be “digested” before the information becomes useful. A rare individual can look at a sheet containing numerous columns containing vast amounts of numbers and quickly analyze most of the information. The majority of us need additional help in unraveling the information so a thorough knowledge is gained to understand if the process is out of control.


While it is simple and easy to compare one number with another number, such comparisons are usually limited. They are limited because of the amount of data used and needed for the comparison is relatively weak in importance. Simple comparisons are limited because the numbers are subject to biological variation that is inevitable in our current production systems. Since the two values being compared are subject to this variation, it is always difficult to determine just how much importance should be placed on what is being indicated. In other words, is the difference due to real change in your current process, or is it biological noise.

Nevertheless, site managers and production managers use the weekly and monthly reports to run the unit that they are overseeing. This can be a breeding unit, or a grow finish unit. For the purpose of this article, let us assume that the production record information is accurate. (Accuracy continues to improve as data can be input in the barn on hand-held devices, etc.) There are a few constraints with each particular data system that need to be factored in if the constraints impact the accuracy of the information that was entered. Most records systems have made it simple to compile data and production parameters, so viewing the data is not a constraint.


A number of years ago, Dr. Polson saw the need to motivate many of us into a better understanding on how to look at production records over time. One of his key points was that data is generally collected with the thought that it will motivate people into an action plan. However, unless the potential signals within the data are separated from biological or probable noise, the actions taken may be totally inconsistent with the data. Thus the proper use of data requires that you have simple and effective methods of analysis which will properly separate potential signals from biological noise.

Methods for statistical process measurement developed by W. A. Shewhart (1) and described by others (2), can be utilized to analyze production records. Chart “signals” indicate a significant deviation from the range of values expected for the production process that you are measuring and identifying. Decision rules (3) helps illustrate the difference between biological noise and real trends.

Statistical process control (SPC) is an effective method of monitoring a process through the use of control charts. Control charts enable the use of objective criteria for distinguishing background biologic variation from events of significance based on statistical techniques. Much of its power lies in the ability to monitor both process center (mean of the data points) and its variation about that center, by collecting data from samples at various points within the process. Variations in the process that may affect the quality of the end product (what we are measuring) can be detected and corrected. With its emphasis on illustrating the variation around the mean, SPC charting has a distinct advantage over other charting methods.


SPC charting indicates when an action should be taken in a process, but it also indicates when NO action should be taken. Chart 1 is a chart of average daily gain for a finisher unit in 2009. The mean for all of the data points is 1.96 average daily gain (ADG) for this time period. This is a very good production level for an average, but one can quickly see that a seasonal dip in performance is documented by Rule 4. Rule 4 is a weaker signal that the “process”, in this case ADG has been affected. Further diagnostic work or investigation is needed to better understand the risk factors involved, but nutritional quality and environmental challenges are two common ones.

Statiscal process control charting has become a valuable tool for practitioners to use to illustrate when intervention is needed and justified. In addition, it takes reports that can be confusing and provides a method of illustrating production levels reached before and after implementation of a change. One of the best uses is to monitor the production level after a change has been implemented to realize if the change was correct and proper (Chart 2). This type of charting has been valuable to illustrate to owners and production staff that their unit is progressing or reaching their farm’s goals.

Chart 1: example of average daily gain for farm x in 2009

Chart 2: Nursery - % mortality 12 month time period


  1. Shewhart, W. A. (1939). Statistical Method from the Viewpoint of Quality Control. New York, New York: Dover Publications, Inc.
  2. Holck, J. T. (1997). Statistical Process Control: Potential applications in the pork industry. Iowa State University Swine Disease Conference, (pp. 31-36). Ames, Iowa.
  3. Wheeler, D. J. (1992). Understanding Statistical Process Control, 2nd Ed. Knoxville, Tennessee: SPC Press.

Dr. Tom Gillespie is the owner and founder of Rensselaer Swine Services. He graduated from Purdue University with a DVM degree in 1979 and initially entered into a mixed animal practice in Illinois before moving to a mixed animal practice in Rensselaer in 1981. After several years of focusing on swine production medicine, he started Rensselaer Swine Services, P.C. in 1991.